CrERF5, an AP2/ERF Transcription Factor, Positively Regulates the Biosynthesis of Bisindole Alkaloids and Their Precursors in Catharanthus roseus

Abstract

Catharanthus roseus contains a variety of monoterpenoid indole alkaloids (MIAs), among which bisindole alkaloids vinblastine and vincristine are well-known to have antitumor effects and widely used in clinical treatment. However, their contents in C. roseus is extremely low and difficult to meet market demands. Therefore, it is of great significance to study the transcriptional regulation mechanism of MIAs biosynthesis for high yielding of bisindole alkaloids in C. roseus. Studies have shown that MIAs biosynthesis in C. roseus has complex temporal and spacial specificity and is under tight transcriptional regulation, especially bisindole alkaloids. In this study, an AP2/ERF transcription factor CrERF5 was selected by RNA-seq of C. roseus organs, and its full-length sequence was cloned and characterized. CrERF5 responds to both ethylene and JA signals and is localized in the nucleus. CrERF5 could activate the transcriptional activity of the TDC promoter. Transient overexpressing CrERF5 in C. roseus petals caused a significant increase of the expression levels of key genes in both the upstream and downstream pathways of MIAs biosynthesis while silencing CrERF5 resulted in a decrease of them. Accordingly, the contents of bisindole alkaloids anhydrovinblastine and vinblastine, monoindole alkaloids ajmalicine, vindoline, and catharanthine were strongly enhanced in CrERF5-overexpressing petals while their contents decreased in CrERF5-silenced plants. These results suggested that CrERF5 is a novel positive ethylene-JA-inducible AP2/ERF transcription factor upregulating the MIAs biosynthetic pathway leading to the bisindole alkaloids accumulation.

Keywords: AP2/ERF transcription factor; Catharanthus roseus; MIAs biosynthesis; bisindole alkaloids; transcriptional regulation.

FIGURE 1

Identification and sequence analysis of CrERF5. (A) Hierarchical cluster analysis of AP2/ERF transcription factors in C. roseus flower, leaf, stem, and root. The average linkage hierarchical clustering with Pearson correlation was used. The color scale at the top represents the value of transformed reads per kilobase per million mapped reads. (B) Phylogenetic analysis of CrERF5 and other known AP2/ERF TFs from five plant species. Arabidopsis (At), Solanum lycopersicum (Sl), Artemisia annua (Aa), Nicotiana tabacum (Nt), and Ophiorrhiza pumila (Op). The tree presented here is a neighbor-joining tree based on amino acid sequence alignment.

FIGURE 2

Subcellular localization and expression of CrERF5. (A) Subcellular localization of CrERF5 in tobacco leaves. Bars = 30 μm. (B) Relative expression of CrERF5 in flower, younger leaf (yl), older leaf (ol), stem, and root of C. roseus. (C) The expression pattern of CrERF5 in C. roseus leaves after treatment with 0.1 mM MeJA, 0.1 mM ETH, and sterile water (ck). The values of hormone treated samples were normalized to the ck samples at each time point, and the ck samples were normalized to themselves. The C. roseus N2227 gene was used as an internal control. The error bars represent the means ± SD (standard deviation) from three replicates.

FIGURE 3

Transient Dual-luciferase assay of CrERF5 with MIA pathway genes (TDC, DXS1, SGD, and CrPRX1) promoters in tobacco. The relative LUC activity was normalized to the reference renilla (REN) luciferase. The error bars represent the means ± SD from three biological replicates, and asterisks indicate statistically significant differences compared with the controls. ^*P < 0.05.

FIGURE 4

Expression of CrERF5 and MIA pathway genes in CrERF5-silencing C. roseus lines. (A) Iridoid pathway genes; (B) Vindoline pathway genes; (C) Indole and MIA pathway genes. The error bars represent the means ± SD from three biological replicates, and asterisks indicate statistically significant differences compared with the controls. ^*P < 0.05.

FIGURE 5

Contents of secologanin, ajmalicine, catharanthine, vindoline, anhydrovinblastine, and vinblastine in CrERF5-silencing C. roseus lines. The error bars represent the means ± SD from three biological replicates, and asterisks indicate statistically significant differences compared with the controls. ^*P < 0.05.

FIGURE 6

Expression of CrERF5 and MIA pathway genes in C. roseus flower petals transiently overexpressing CrERF5. (A) Iridoid pathway genes; (B) Vindoline pathway genes; (C) Indole and MIA pathway genes. The error bars represent the means ± SD from three biological replicates, and asterisks indicate statistically significant differences compared with the controls. ^*P < 0.05.

FIGURE 7

Contents of secologanin, ajmalicine, catharanthine, vindoline, anhydrovinblastine, and vinblastine in C. roseus flower petals transiently overexpressing CrERF5. The error bars represent the means ± SD from three biological replicates, and asterisks indicate statistically significant differences compared with the controls. ^*P < 0.05.

FIGURE 8

A scheme of CrERF5 regulation on MIA biosynthetic pathway. Genes regulated by CrERF5 are boxed in pink, respectively. Genes not regulated by CrERF5 are boxed in gray. MIAs upregulated by CrERF5 are in pink.